5 Money Market Third Quarterly Report for FY4 After the reversal of the December 23 upsurge in short-term rates, the market entered a period of relative stability. While it continued to expect a modest increase in demand for government borrowing, it was also concerned about a number of related issues. These included the steady increase in domestic inflation, the rise in international interest rates and the narrowing current account surplus, the SBP had successfully contained the expectations of a sharp rise in interest rates by allowing only a very gradual increase in T-bill cut-offs (see Figure 5.1). Figure 5.1: -month Yields Auction cut-off Repo rate 2.9 2.5 2.1 1.7 1.3.9 Cut-off trend 1-Jul-3 3-Jul-3 19-Aug-3 8-Sep-3 28-Sep-3 18-Oct-3 7-Nov-3 27-Nov-3 17-Dec-3 -Jan-4 2-Jan-4 15-Feb-4 -Mar-4 2-Mar-4 15-Apr-4 5-May-4 25-May-4 Figure 5.2: T-bill Auction Cut-offs 3-month -month 12-month 3. As evident in Figure 5.2, short-term rates only saw a 2.5 modest increase during Q3- FY3. However, as in Q2-2. FY3, market expectations 1.5 were jolted by the April 24 announcement of an 1. unexpectedly large PIB auction. This, together with an unseasonal acceleration in inflation, as well as an incremental narrowing of the current account 1 revived expectations of a large movement in interest rates. Accordingly, term rates climbed strongly April 24 onwards. 2 3-Apr-3 11-May-3 18-Jun-3 2-Jul-3 2-Sep-3 1-Oct-3 17-Nov-3 25-Dec-3 1-Feb-4 1-Mar-4 17-Apr-4 25-May-4 1 This owed to both a jump in imports as well as the termination of the Saudi Oil Facility (highlighted in the April 24 release of the end-march 24 balance of payments data). The latter, in particular, implied a dual negative impact: the lower external account surplus suggested a decline in rupee injections due to lower SBP forex purchases, while simultaneously increasing the government s reliance on domestic funding to finance the fiscal deficit. 2 By end-may 24, the rise was due to factors related to the PARCO swap transaction (this floating rate borrowing was linked to the average market prior to the transaction date). 47
The State of Pakistan s Economy Box 5.1: Karachi Interbank Offered Rate (KIBOR) KIBOR is defined as the average rate (ask side), for the relevant tenor, as published on the Reuters page KIBOR or as published by the Financial Markets Association of Pakistan in case the Reuters page is unavailable. Initially, introduced in September 21, KIBOR was only used as a reference rate for interbank money market (for clean lending). However, to promote the culture of floating rate lending and make the mechanism transparent both for lender as well as borrower, KIBOR was also introduced as a reference rate for corporate lending in February 24. Initially KIBOR of one-month; three-month and six-month tenors would be used as benchmark for all-corporate lending in the local currency. Subsequently, It was successfully extended to one year on March 31, 24 and would be extended to three years by December 31, 24. KIBOR will not be applicable on (1) export finance scheme (2) consumer financing and SME lending (3) overdrafts and running finance facilities existing before January 31, 24 (4) term finance certificates/ commercial papers approved by SECP or submitted to any stock exchange before January 31, 24; and (5) all term loans with agreements executed before January 31, 24. However, in case of re-pricing, KIBOR will be applicable in available tenors. 5.1 Term Structure of Interest Rates During Q3-FY3, there was little change in the longer end of the yield curve, other than its successful extension with the inaugural issue of 15- and 2-year Treasury bonds (PIBs) in January 24 (see Figure 5.3). However, there were some movements in shorter tenors, in line with the developments discussed earlier. Initially, the yield curve pivoted down to become steeper following the post-december 23 fall in short-term rates and Figure 5.3: Yield Curves 24-Dec-3 -Jan-4 3-Apr-4 11-Jun-4 1 per annum 8 4 2 3m m 1y 3y 5y 1y 15y 2y maturity thereafter flattened only slightly by April 23 (largely mirroring the very gradual rise in the T-bill auction cut-offs). The stability of the long-term rates owed, in part, to the lack of large PIB auctions during the period. This lowered pressure on PIB prices and complemented the SBP s policy of raising short-term rates only very gradually. However, the 48
Third Quarterly Report for FY4 market s interest rate outlook underwent a transformation with the unexpected large PIB issue announced in April 24, which suggested a strong government appetite for borrowings. The resulting expectation of a sharper rise in the interest rates was then reinforced by relatively larger upward shift of the acceptance cutoffs in the next two T-bill auctions (as inflationary concerns grew). The scale of the shift in market expectations by mid- May 24 is evident from changes in the profile of the bids in T-bill auctions. For example, in the 3-month T- bill auction for June 9, 24, the lowest bid was substantially above the highest bid of the previous auction (see Figure 5.4). These developments clearly highlight the importance of a greater coordination between the SBP (that sets targets for T-bills auctions) and the Finance Ministry (which sets targets for PIBs), in managing the interest rate expectations in the economy. This becomes even more important in the case of Pakistan given that during FY4 (to date), net government market borrowings through PIBs have been significantly higher than through T-bills. Figure 5.4: 3-month T-bill Auction Bid Pattern (high, low, cut-off) 4. 3.5 3. 2.5 2. 1.5 1. 2-Dec-3 1-Jan-4 25-Jan-4 9-Feb-4 24-Feb-4 1-Mar-4 25-Mar-4 9-Apr-4 24-Apr-4 9-May-4 24-May-4 8-Jun-4 Table 5.1: Secondary Market Trading 3m m 12m PIB Combined Q1-FY4 Total 22. 25.1 77.9 7.2 1,815.2 Average.3 3.4 1.2 1. 23.9 Max 3. 34.9 39.8 18.1 4.4 Q2-FY4 Total 2.8 21.7 989.8 1,151.9 2,379.1 Average.3 3. 13. 15.8 32. Max 3.5 8.1 28.1 52.7 52.7 Q3-FY4 Total 9.7 248. 7.2 1,31.9 2,11.4 Average 1.2 3. 11.1 15.2 3.7 Max 9.2 52.9 21.8 2.8 81.4 April 24 Average 2. 2.3 12.2 15.3 31.8 5.2 Trading Volumes 3 As seen from Table 5.1, secondary market trading in government securities has averaged over Rs 3 billion in Q2 and Q3 of FY4, which is substantially higher than for the corresponding figures for FY3. 3 Reported volumes are based on simple aggregation of SGLA movements. 49
The State of Pakistan s Economy Another very significant development in recent quarters is the increased interbank trading in PIBs. As seen in Figure 5.5, the increased trading in PIBs appears to be co-related with the rise in outstanding stock of these instruments with the banking sector. 5.3 SBP Market Support and Rupee Interventions As shown in Figure 5., the frequency and direction of OMOs during Jan-Apr 24 was in sharp contrast to those in the corresponding period of FY3. In specific terms, there was no OMO Jan-Apr 23 compared with seven OMOs during the corresponding period of the current year. These OMOs (Jan-Apr 24) were geared to stem the excessive growth in monetary aggregates (M2 and reserve money). As shown in Figure 5.7 and Table 5.2, barring a few episodes of discounting, the market remained relatively more liquid during Jan-Apr 24 as compared with Q2- FY4 and the corresponding period last year. As reported in Table 5.3, the average level of overnight rates showed a decline of Figure 5.5: Banks' PIB Holdings and Trading Holding as % of DTL Trading share (RHS) Figure 5.: Open Market Operations Injection Absorption Frequency (RHS) 11 1-9 8 7 78 52 2 Aug-2 Oct-2 Dec-2 Feb-3 Apr-3 Jun-3 Aug-3 Oct-3 Dec-3 Feb-4 Apr-4 Jul-2 Oct-2 Jan-3 Apr-3 Jul-3 Oct-3 Jan-4 Apr-4 Figure 5.7: Discounting and Overnight Dis. O/N rate (RHS) DR (RHS) 2 1 12 8 4 1-Oct-3 1-Oct-3 31-Oct-3 15-Nov-3 3-Nov-3 15-Dec-3 3-Dec-3 14-Jan-4 29-Jan-4 13-Feb-4 28-Feb-4 14-Mar-4 29-Mar-4 13-Apr-4 28-Apr-4 4 2-54 48 42 3 3 24 8 4 2 number 1 per annum 5
Third Quarterly Report for FY4 Table 5.2: Activities at Discount Window No. of visits Total discounting Average per visit FY2 FY3 FY4 FY2 FY3 FY4 FY2 FY3 FY4 1st Quarter 39 1-11.5 144.1-4.1 9. - 2nd Quarter 57 32 3 33.2 325.3 1.9 5.9 1.2 3. January 5 1 1 17.4 14.2 1.4 3.5 14. 1.4 February 8 1 2 12. 2.5 8.3 12.7 2.5 4.2 March 1-1.4-1.4 - - 3rd Quarter 14 11 3 129.7 142.7 9.7 9.3 13. 3.2 around 7 bps in Q3-FY4 compared with the last quarter, reflecting relative liquidity ease in the market. However, volatility increased despite more frequent market intervention by the SBP. 5.4 Treasury-bill Auctions As shown in Figure 5.8, the targets set for T-bill auctions during January-April 24 were typically higher than maturities. Moreover, other than a single auction (January 22, 24) all attracted substantial interest and the amounts accepted were typically slightly higher than the amounts maturing, indicating the ample liquidity available with banks. Table 5.3: Volatility of Overnight Rates Std Dev Average Coef. of Var Oct-3 1.9 2.2.8 Nov-3 1.7 2.2.8 Dec-3 2.3 2.2 1. Q2-FY4 2. 2.2.9 Jan-4 2.2 1.8 1.2 Feb-4 2.8 1.9 1.5 Mar-4.8 1..8 Q3-FY4 2.1 1.5 1.4 Apr-4 2.4 2.7.9 Figure 5.8: T-bill Auctions (net of maturity) Target Offer Accepted 4 2-2 -4 - However, it is significant to note that for most auctions the acceptance cut-off lay towards the lower end of the bid-spread. In other words, despite the apparent liquidity, banks continued to demand a larger increase in yields than what the SBP was willing to offer. 8-Jan-4 The ability of banks to do so, in turn, depended on a number of factors, these included the low opportunity cost of uninvested funds (T-bill auction yields were 22-Jan-4 5-Feb-4 19-Feb-4 4-Mar-4 18-Mar-4 1-Apr-4 15-Apr-4 29-Apr-4 13-May-4 27-May-4 51
The State of Pakistan s Economy marginally higher than the average overnight rate), as well as the investment opportunities offered by large PIB issues. 5.5 Pakistan Investment Bond (PIB) Auctions Contrary to initial market expectations, Q3-FY3 witnessed only one small (Rs billion) PIB auction 12 (marking the launch of the 15- and 2-year PIB). As 9 shown in Figure 5.9, the January 24 auction of 15- & 2-year bonds generated a lot of interest especially 3 amongst non-bank institutions. The amount on offer was almost twice the announced target, and not surprisingly the acceptance cut-off was at a premium (see Table 5.4). However, it was the announcement of a Rs 4 billion Jumbo issue, which was to be issued in two tranches (Rs 25 billion for April 24 and the remainder in May 24), which had a significant impact on market expectations.. Figure 5.9: PIB Auction (15 and 2 years) Target Offered Accepted Table 5.4: PIB Auction (15 & 2 years maturity) Auction held on January 19, 24 Acceptance () Non Short selling Total titi 15-year.1.3 3.3 2-year.2.4 3.2 Total.2.7.5 Interest rates (in ) 15-year 2-year Total Coupon Cut-off W. Average 15-year 9. 7.8 7.7 As seen in Table 5.4, the April 24 offering (first tranche of the Jumbo issue) reflected the heavy market demand for long-term bonds. 2-year Price (in Rupees) 15-year 2-year 1. Highest 112.7 113. 8.8 Lowest 15. 15.2 8.7 Range 7.7 8.4 All tenors were oversubscribed Note: Totals may not tally due to separate rounding-off and the tenor-wise targets were met comfortably. Despite a slight decline in yields of 3- and 5-year bonds, the yield on the 1-year bond increased by 27 bps. Intuitively, it seems that this difference in the direction of change in the 1-year bond yield relative to 52
Third Quarterly Report for FY4 that other tenors is largely due to the significantly larger amounts offered in the longer tenor instrument. 4 However, the May 24 PIB auction (second tranche of the Jumbo issue) saw a more than 8 bps rise in PIB cut-off yields (see Table 5.5). This meant that the yield curve became steeper (as this rise was greater than the rise in cut-off yields in the earlier T- bill auction). This steepening yield curve, in turn, (1) put an upward pressure on shortterm interest rates, and (2) intensified banks interest in long-term government paper (ironically, this came precisely when the SBP was warning banks on the risks of very high PIB holdings). 5 As shown in Figure 5.1, the share of commercial banks in the total stock of outstanding PIBs has declined in the period January-April 24. It shows that during this period, fresh issues were largely purchased by non-bank institutions. While this development is quite welcome, the PIB holdings of Table 5.5: PIB Auction (3, 5 & 1 years maturity) auction held on April 29, 24 3-year 5-year 1-year Total Target 3. 7. 15. 25. Offered 7.7 15.7 23.7 47.1 Accepted 3. 8.3 15. 2.3 Cut-off (price) Rupees 1.3 19.2 111. - Coupon. 7. 8. - Cut-off (yield) 3.8 4.9.5 - auction held on May 29, 24 Target 2. 3. 1. 15. Offered 3. 4.3 1.4 17.7 Accepted 1.9 3. 1. 14.9 Cut-off (price) Rupees 14.5 17.1 14.4 - Coupon. 7. 8. - Cut-off (yield) 4.4 5.4 7.4 - Figure 5.1: Banks' PIB Holdings Share Amount (RHS) 8 7 5 4 3 2 Jun-1 Sep-1 Dec-1 Mar-2 Jun-2 Sep-2 Dec-2 Mar-3 Jun-3 Sep-3 Dec-3 Mar-4 19 1 13 1 7 4 1 4 One interesting explanation conjectured for the higher supply in 1-year would tend to increase its yield and thereby lowering its differential with the DSC yields. This, in turn, would permit the government to avoid a (quite unpopular) large cut in NSS rates (which are issued at a small premium on PIB yields). 5 As discussed earlier, this highlights the need the for greater coordination between T-bill and PIB auctions, in terms of timing and size of issuances and their probable impact on interest rate movements. 53
The State of Pakistan s Economy banks remain uncomfortably high, given the potential for large capital losses in case of a significant upward shift in interest rates. 54